license: apache-2.0
base_model: unsloth/Qwen3.5-2B
tags:
- ascii
- vision
- art
- unsloth
- image-to-text
datasets:
- PinkPixel/ASCII-Art
pipeline_tag: image-to-text
library_name: transformers
✨ ASCII Machine ✨
ASCII Art Generation Model
🚀 Model Overview
ASCII Machine is a specialized vision-language model designed for high-quality ASCII art generation and interpretation. Based on the Qwen3.5-2B architecture, it has been fine-tuned to excel at transforming visual concepts into intricate text-based masterpieces.
- Model Name: PinkPixel/ASCII-Machine
- Base Model: Qwen3.5-2B
- Training Platform: Unsloth Studio
- Specialization: ASCII Art Generation & Vision-to-ASCII tasks
🎨 Dataset
This model was trained on the PinkPixel/ASCII-Art dataset, a curated collection of thousands of diverse ASCII artworks across various styles and subjects.
💬 Example Usage
User:
Can you create a small ASCII art of a cat?
ASCII Machine:
^__^
(oo)\_______
(__)\ )\/\
||----w |
|| ||
(Note: Actual output quality may vary based on prompt complexity and artistic style requested)
👁️ Vision Capabilities
As a vision-enabled model, ASCII Machine can "see" images and describe them or convert them directly into ASCII representations. Please note that this uses a cutting-edge architecture; while the full safetensors version supports vision natively in compatible environments, third-party support in tools like llama.cpp or LM Studio may still be in development.
🛠️ Installation & Usage
You can use this model directly with the Hugging Face transformers library:
from transformers import AutoProcessor, AutoModelForVision2Seq
model_id = "PinkPixel/ASCII-Machine"
processor = AutoProcessor.from_pretrained(model_id)
model = AutoModelForVision2Seq.from_pretrained(model_id)
"Dream it, Pixel it"